| Abstract|| |
The aim of our study was to find out a new indicator with a higher specificity level than prostate prostate-specific antigen (PSA) in order to achieve a better selection of patients who will undergo prostate biopsy. Trans-rectal ultrasound-guided prostate biopsy was performed in 135 patients who had elevated PSA level and/or palpable nodule on digital rectal examination. The PSA level was ≤10 ng/mL in 81 patients and >10 ng/mL in 54 patients. We designed a new formula consisting of prostate volume, patient's age, and free prostate specific antigen. Its resultant was defined as prostate biopsy index and was compared with the most currently used parameters. Histology results yielded prostate gland malignancy in 40 (30%) patients. Our new index differed significantly between the malignant and the non-malignant patient categories (P = 0.01). The ROC curve analysis at different specificity and sensitivity levels (85%, 90% and 95%) and regarding the area under the curve (AUC), our new index was significantly better than the other studied parameters (P = 0.001). Additionally, the AUC in patients with a PSA level ≤10 ng/mL and between 10.1 and 20 ng/mL was 0.75 and 0.78, with a sensitivity of 91% and 83% and a specificity of 24% and 73%, respectively, at a cut-off point of 1.7. The overall sensitivity and specificity at the same point were 80% and 41%, respectively. In conclusion, the performance of our new index was superior to all other evaluated parameters. At 83% sensitivity with a cut-off point of 1.7, 63.5% of the performed biopsies could have been avoided in patients with a PSA level between 10.1 and 20 ng/mL.
|How to cite this article:|
Haroun AA. New indicator for prostate gland biopsy when malignancy is in question. Saudi J Kidney Dis Transpl 2011;22:61-6
|How to cite this URL:|
Haroun AA. New indicator for prostate gland biopsy when malignancy is in question. Saudi J Kidney Dis Transpl [serial online] 2011 [cited 2020 May 26];22:61-6. Available from: http://www.sjkdt.org/text.asp?2011/22/1/61/74349
| Introduction|| |
By the end of the last century, prostate specific antigen (PSA) had become a main predictor for prostate cancer. Its discriminatory power between malignant and benign prostatic diseases is poor.
Several PSA derivatives have been developed in order to improve cancer detection, whose results and thresholds were largely variable, and did not find a wide acceptance. 
Prostate volume was found to be a good predictor for prostate cancer by some authors. , It was considered as an important intermediary between PSA levels, prostate cancer and benign prostatic hyperplasia (BPH).  Other authors did not find any evidence in their study that allowed detection of prostate malignancy based on gland size alone. 
However, its measurement by using the transrectal approach had been described to be in-accurate when compared with prostatic weight at prostatectomy. 
Trans-rectal ultrasound (TRUS)-guided prostate biopsy is uncomfortable to patients, is time consuming and may be overprescribed. For these reasons, we attempted, in this retrospective study, to find out a new indicator for prostate biopsy with a better specificity level than the currently used parameters, which could permit a better patient selection.
We evaluated the prostate gland volume, patient's age, free prostate specific-antigen (fPSA) serum level, total prostate-specific antigen (t-PSA), free to total PSA ratio (%f-PSA) and prostate-specific antigen density (PSAD).
| Materials and Methods|| |
The clinical reports of 164 patients who underwent TRUS-guided prostate gland biopsy over a 24-month period were reviewed. The indications of prostate biopsy were elevated PSA level (>4 ng/mL) and/or palpable nodule on digital rectal examination.
Complete records were found for 135 patients. Their mean age was 67 years, ranging from 50 to 83 years.
Blood samples were usually taken at least three weeks before TRUS biopsy, stored at 4°C and examined during the 1st 24 hours. The total and free PSA values were determined using the ARCHITECT Total PSA and ARCHITECT Free PSA assays (Abbott, Ireland).
TRUS-guided prostate biopsy was performed by using a 7.5 MHz endocavitary probe. The prostate gland volume was measured according to the prolate ellipsoid formula (0.52 × length × width × height).
All TRUS biopsies were performed by the same radiologist (author), permitting a consistent method of prostate volume estimation, and same sampling technique for all patients. Eight biopsy samples were obtained from each patient and an additional two targeted cores from any hypoechoic suspicious area in the peripheral zones was also performed. According to the prostate gland volume, the patients were stratified into four groups: group I with prostate volume between 20 and 40 mL, group II with prostate volume between 41 and 60 mL, group III with prostate volume between 61 and 80 mL and group IV with prostate volume >80 mL. The patient's age, t-PSA, f-PSA, %f-PSA, PSAD and histology results were determined for each patient in each group.
We designed a simple formula consisting of three parameters: prostate volume, patient's age and f-PSA serum level. The prostate volume was divided by patient's age and then multiplied by the f-PSA serum level: (prostate volume/patient's age) × f-PSA. The resultant of this formula was defined as the prostate biopsy index (PBI).
| Statistical Analysis|| |
The correlation between different parameters was assessed using Pearson's correlation coefficient. A two-sample, two-sided Student's ttest was used to assess differences between malignant and benign conditions.
An analysis of variance (ANOVA) was carried out to assess differences in mean levels of the parameters between the groups (I-IV).
ROC curves for the investigated parameters were generated. AUC and the associated Pvalue were determined. Sensitivity, specificity and predictive values of our new index (PBI) and other investigated parameters were also measured and compared. All analyses were carried out using SPSS version 15.
| Results|| |
Histology results yielded 40 patients (30%) with prostate cancer, 86 patients (64%) with benign prostate diseases and nine patients (6%) with no abnormal findings.
Four patients (3%) had a PSA level ≤4 ng/ mL, 77 patients (57%) were in the gray area (PSA level 4.1-10 ng/mL), 39 patients (29%) had a PSA level of 10.1-20 ng/mL and 15 patients (11%) had a PSA level of >20 ng/mL. The incidence of malignancy in these patients was 1.5%, 15%, 9% and 4.5%, respectively. The mean prostate volume, t-PSA and %f-PSA were 58 ± 27 mL, 12.2 ± 10.8 ng/mL and 0.18 ± 0.1, and the mean f-PSA and PSAD were 1.9 ± 1.35 ng/mL and 0.25 ± 0.27, respectively, irrespective of the histological results. The detailed clinical characteristics of the entire study group are summarized in [Table 1].
|Table 1: Clinical characteristics of the cohort according to prostate volume groups and histology results in 135 patients.|
Click here to view
The only parameters that were statistically significantly different between the malignant and the benign categories were prostate volume (46.7 ± 23 and 62.5 ± 28 mL, respectively, P = 0.001) and PSAD (P = 0.01).
Within the volume groups, all the evaluated parameters in group II only were significantly different, except %f-PSA between the malignant and the non-malignant conditions (P = 0.03, 0.005, 0.005 and 0.003, for patient's age, t-PSA, PSAD and f-PSA, respectively). There was a stronger correlation between prostate volume and patient's age in the malignant than in the non-malignant conditions (r = 0.32, 0.15; P = 0.04 and 0.16, respectively). There was also a positive correlation between prostate volume and f-PSA (r = 33, 0.30, P = 0.06 and 0.001, respectively). The mean value of our PBI in the malignant and the non-malignant categories was 1.22 ± 1.5 and 2 ± 1.8, respectively (P = 0.01).
The ROC curve showed that AUC for our PBI was higher than the rest of the studied parameters (P = 0.001) [Table 3], [Figure 1].
|Figure 1: ROC curves show AUC for PBI (0.78), t-PSA, %f-PSA and PSAD in patients having a t-PSA|
level between 10.1 and 20 ng/mL.
Click here to view
At sensitivity levels of 85%, 90% and 95%, our PBI also performed much better [Table 2] and [Table 3].
|Table 2: Specificity of PBI and other variables at different sensitivity levels.|
Click here to view
|Table 3: Specificities, predictive values and cut-off points with corresponding area under the curve for|
PBI, PSA and its derivatives at a given 90% sensitivity.
Click here to view
When the efficiency of our index at the cutoff point of 1.7 was tested in patients according to PSA levels, we found that the AUC was 0.75 (P = 0.000), 0.78 (P = 0.006) and 0.74 (P = 0.12) for patients with PSA levels ≤10 ng/mL, 10.1-20 ng/mL and >20 ng/mL, respectively. The sensitivity levels were 91%, 83% and 30%, with a corresponding specificity of 24%, 73% and 30%, respectively. In patients with t-PSA levels of 10.1-20 ng/mL, the AUC for t-PSA, %f-PSA and PSAD was 0.58, 0.7 and 0.31, respectively, [Figure 1].
| Discussion|| |
Prostate volume is dependent on multiple factors. Some studies demonstrated a relationship between prostate volume and other variables such as race, ,, body weight  and sex hormone-binding globulin level. 
A positive relationship between prostate volume and patient's age is observed in malignant cases. Horninger et al.  found a slight influence of age on cancer detection rates, which was higher in the elderly. Kobayashi et al.  also suggested that the influence of prostate volume on prostate cancer detection was stronger in elderly men. It had been reported that prostate volume was superior to PSA for predicting positive results of a prostate biopsy. ,, How ever, the published reports regarding PSAD were largely variable and non-determinant. In our study, PSAD had the least AUC among the other investigated parameters in patients with t-PSA 10.1-20 ng/mL.
The incidence of prostate cancer in small prostates (≤40 mL) was higher than that in large prostates (>80 mL), which was 50% versus 9%, and this was similar to that reported in several reports. ,,
Some authors reported that the tumor size is smaller in large glands,  and the reason for the lower detection rate in larger glands could be related to sampling error from such fewer core samples. ,, Abdel-Khalek et al.  stated that men with one negative extended prostate biopsy have a 19% probability of having cancer on repeating the procedure, and this probability is 23% in men with an initial negative sextant biopsy. In another study, these rates were 10% and 50%, respectively. 
The number of biopsy cores obtained in our patients was 8-10 and the incidence of prostate cancer was 30%, and that was within the reported range (22-37%); some of these reports used ≥10 cores ,,, and others used the sex tant biopsy technique. , In one recent study,  sextant biopsy technique plus site-specific biopsy was performed in 1644 patients, wherein the prevalence rate of prostate cancer was 43%, which was much higher than that reported in many other reports using a greater number of biopsy cores.
Hammerer and Huland  reported an incidence of cancer in 45% using a sextant biopsy technique.
Other studies did not also demonstrate a significant difference in cancer detection rate with increasing number of biopsy samples. ,, The results of all these reports incite to look for additional factors other than tumor size and grade that may affect the cancer detection rate in large glands.
Our results demonstrated that f-PSA correlated significantly with prostate volume. Its mean value was statistically higher in malignant than in benign conditions for patients with a prostate volume between 40 and 60 mL (P = 0.003).
Because of a variability of parameters that either affect the prostate gland or reflect the prostate gland pathology, and due to the wide variation of reported results concerning these parameters, we designed a formula based on the most common variables that are usually encountered in prostate gland diseases, depending on our results and on the published reports in this entity. Therefore, we found that the prostate volume, patient's age and f-PSA serum level were the most important factors. We incorporated these parameters to obtain an index with a certain cut-off point. The mean value of our new index was significantly lower for the positive biopsy group than for the negative biopsy group (P = 0.01).
At different sensitivity and specificity levels (85%, 90% and 95%), the performance of our new index was much better than with PSA and its derivatives.
The value of a tumor marker test is determined by its specificity level, and the characteristic of a screening test described by specificity is more important in determining the predictive value of a positive test than is sensitivity. In order to have a better compromise between sensitivity and specificity, we suggest a cut-off point at 1.7 for our index, which allows a sensitivity of 80% and a specificity of 41% irrespective of the PSA level.
The limitations of our study included its retrospective nature, relatively small sample size population and lack of follow-up of patients and their respective PSA levels over time. We hope that our results open the gate for other researchers to continue the work at our index, particularly on measuring the prostate volume by a transabdominal approach, which would spare patient's discomfort, time and costs.
The discriminatory power of our new index (PBI) between malignant and non-malignant conditions was statistically superior to PSA levels and their derivatives. In addition, its performance according to ROC curve analysis was much better at different sensitivity and specificity levels. Therefore, we believe that our index will be much more helpful for clinicians than other currently used parameters when a biopsy decision is required, particularly in patients with a PSA level between 10.1 and 20 ng/mL, where 63.5% of the performed biopsies could have been spared in keeping a sensitivity of 83%. Multicenter studies on a larger number of patients with different races are warranted.
| References|| |
|1.||Horninger W, Cheli CD, Babaian RJ, et al. Complexed prostate-specific antigen for early detection of prostate cancer in men with serum prostate-specific antigen levels of 2 to 4 nanograms per milliliter. Urology 2002;60 (Suppl4.1):31-5. |
|2.||Al-Azab R, Toi A, Lockwood G, Kulkarni GS, Fleshner N. Prostate volume is strongest predictor of cancer diagnosis at transrectal ultrasound-guided prostate biopsy with prostatespecific antigen values between 2.0 and 9.0 ng/ml. Urology 2007;69(1):103-7. |
|3.||Kobayashi T, Mitsumori K, Kawahara T, Nishizawa K, Ogura K, Ide Y. Prostate gland volume is a strong predictor of biopsy results in men 70 years or older with prostate-specific antigen levels of 2.0-10.0ng/ml. Int J Urol 2005;12(11):969-75. |
|4.||Pinsky PF, Kramer BS, Crawford ED, et al. Prostate volume and prostate-specific antigen levels in men enrolled in a large screening trial. Urology 2006;68(2):352-6. |
|5.||Uzzo RG, Wei JT, Waldbaum RS, Perlmutter AP, Byrne JC, Vaughan ED. The influence of prostate size on cancer detection. Urology 1995;46(6):831-6. |
|6.||Loeb S, Han M, Roehl KA, Antenor JA, Catalona WJ. Accuracy of prostate weight estimation by digital rectal examination versus transrectal ultrasonography. J Urol 2005;173(1):63-5. |
|7.||Lee SE, Chung JS, Han BK, et al. Relationship of prostate-specific antigen in Korean men with biopsy-proven benign prostatic hyperplasia. Urology 2008;71(3):395-8. |
|8.||Chang YL, Lin AT, Chen KK, et al. Correlation between serum of prostate-specific antigen and prostate volume in Taiwanese men with biopsy proven benign prostatic hyperplasia. J Urol 2006;176(1):196-9. |
|9.||Fowler JE, Bigler SA, Kilambi NK, Land SA. Relationships between prostate-specific antigen and prostate volume in black and white men with benign prostatic hyperplasia. Urology 1999;53(6):1175-8. |
|10.||Ochiai A, Fritsche HA, Babaian RJ. Influence of anthropometric measurements, age, and prostate volume on prostate-specific antigen levels in men with a low risk of prostate cancer. Urology 2005;66(4):819-23. |
|11.||Joseph MA, Wei JT, Harlow SD, et al. Relationship of serum sex-steroid hormones and prostate volume in African American men. Prostate 2002;53(4):322-9. |
|12.||Hong SK, Yu JH, Han BK, et al. Association of prostate size and tumor grade in Korean men with clinically localized prostate cancer. Urology 2007;70(1):91-5. |
|13.||Chen ME, Troncoso P, Johnston D, Tang K, Babaian RJ. Prostate cancer detection, relationship to prostate size. Urology 1999;53(4): 764-8. |
|14.||Al-Ghamdi AM, Lockwood G, Toi A, et al. Extended pattern prostate biopsy does not minimize the volume-grade bias in prostate cancer detection. J Urol 2008;179(4):1332-4. |
|15.||Abdel-Khalek M, El-Baz M, Ibrahim el-H. Predictors of prostate cancer on extended biopsy in patients with high-grade prostate intraepithelial neoplasia: A multivariate analysis model. BJU Int 2004;94(4):528-33. |
|16.||Mian BM, Naya Y, Okihara A, Vakar-Lopez F, Troncoso P, Babaian RJ. Predictors of cancer in repeat extended multisite prostate biopsy in men with previous negative extended multisite prostate biopsy. Urology 2002;60(5): 836-40. |
|17.||Kobayashi T, Kawahara T, Nishizawa K, Ogura K, Mitumori K, Ide Y. Volume-adjusted prostate-specific antigen [PSA] variables in detecting impalpable prostate cancer in men with PSA levels of 2-4ng/ml: Transabdominal measurement makes a significant contribution. BJU Int 2005;95(9):1245-8. |
|18.||Hammere P, Huland H. Systemic sextant biopsies in 651 patients referred for prostate evaluation. J Urol 1994;151(1):99-102. |
|19.||Naughton CK, Miller DC, Mager DE, Ornstein DK, Catalona WJ. A prospective randomized trial comparing 6 versus 12 prostate biopsy cores: Impact on cancer detection. J Urol 2000;164 (2):388-92. |
|20.||Kobayashi T, Nishizawa K, Watanabe J, Ogura K, Mitsumori K, Ide Y. Effects of sextant transrectal prostate biopsy plus additional far lateral cores in improving cancer detection rates in men with large prostate glands. Int J Urol 2004;11(6):392-6. |
|21.||Ung JO, San Francisco IF, Regan MM, De Wolf WC, Olumi AF. The relationship of prostate gland volume to extended needle biopsy on prostate cancer detection. J Urol 2003;169(1):130-5. |
Azmi A Haroun
Department of Radiology, Jordan University Hospital, P.O. Box 460495, 11946 Amman
[Table 1], [Table 2], [Table 3]